Del fundamento del derecho a la igualdad

ACUTE EFFECTS OF INTRADIALYTIC PROTEIN SUPPLEMENTATION

Chronic kidney disease (CKD) is a progressive inflammatory disorder that affects approximately 13% of adults in the U.S, and the prevalence is increasing rapidly1. Advanced CKD requiring dialysis treatment is associated with a variety of metabolic disturbances that increase morbidity and mortality; in addition, protein malnutrition, muscle wasting, bone disorders, and cardiovascular complications are especially common, and these co-morbidities greatly reduce physical function and quality of life in dialysis patients. Furthermore, 2/3 of patients die within 5 years of initiation of long-term dialysis treatment, mostly of cardiovascular disease (CVD)3, and survival has not increased substantially in the past two decades4. New therapeutic approaches are needed to address the many co-morbid conditions associated with advanced kidney disease.

Introduction

Although dialysis treatment procedures and protocols have improved over the years, mortality rates remain elevated in dialysis patients, and many believe protein-energy malnutrition and elevated inflammation are one of the primary reasons for this effect. Protein-energy

malnutrition (PEM) is very common in dialysis patients, with the incidence rate ranging from 25% to 75% in different studies6-9. There are a variety of reasons for this, including poor nutrient intake, physical illnesses affecting gastrointestinal function, protein losses during dialysis, and elevated whole body and skeletal muscle protein catabolism that occurs primarily during dialysis10. Furthermore, PEM and inflammation tend to occur concurrently in dialysis patients, giving rise to the term malnutrition-inflammation complex or syndrome.

Inflammation may also be contributing to PEM by inducing appetite suppression, as high circulating levels of inflammatory cytokines has been shown to increase anorexia in dialysis patients222. This voluntary decrease in nutrient intake further exacerbates the malnutrition inflammation complex syndrome, leading to more inflammation and further declines in

nutritional status. In particular, IL-6 seems particularly important as a marker and mediator of dialysis-associated inflammation. Tripepi and colleagues indicate IL-6, compared to several other inflammatory cytokines, adds the greatest predictive power with regards to all-cause and

cardiovascular mortality in the context of traditional and non-traditional risk factors for dialysis patients232. Furthermore, the dialysis procedure induces an acute increase of IL-6233 during the course of a single treatment that persists into the post-dialysis phase, and other inflammatory markers may be transiently elevated in conjunction with dialysis treatment230.

Anemia of chronic disease, often termed anemia of chronic inflammation, occurs in dialysis patients due to chronic immune activation, infection, and inflammation234. Individuals with advanced CKD very commonly suffer from another type of anemia called anemia of renal insufficiency due to the inability of the kidney to produce erythropoietin. Although the two types of anemia are often present concurrently and share some symptomatic indicators, they are considered to be distinct conditions with different eitologies235. Therefore, inflammation further exacerbates the declining nutritional state of patients who already experience a derangement of nutrient metabolism due to renal insufficiency.

Because protein malnutrition in dialysis patients promotes inflammation, it has been suggested that improving nutritional status may help halt this cycle of inflammation and decreased nutritional status60. If so, this could have a beneficial effect on many CKD co-

morbidities influenced by inflammation, particularly CVD56, 72. In response to these issues, the National Kidney Foundation has recommended an increase in the protein requirement to 1.2 g/kg/day for hemodialysis patients in comparison to the 0.8 g/kg/day recommended for healthy adults and 0.6 g/kg/day recommended for earlier stages of CKD. This recommendation was based on several small prospective nutritional-metabolic studies indicating this intake level is necessary to ensure neutral or positive nitrogen balance in most dialysis patients10. However, most dialysis patients do not consume this recommended amount of protein, and a large multi- center study of dialysis patients showed protein and calorie intake to be significantly lower on dialysis treatment days compared to non-treatment days219. The combination of amino acids lost into the dialysate, suppressed nutrient intake on dialysis days, and the catabolic effects of dialysis treatment suggests that the time immediately prior to initiation of treatment would be most appropriate to administer a protein intervention to potentially offset the negative effects of dialysis.

The purpose of this study was to determine the effects of whey and soy protein supplementation on acute inflammation over the course of a dialysis treatment. While both protein sources will increase substrate availability, they differ in amino acid composition and

presence of other bioactive compounds such as soy isoflavones. We hypothesized protein supplementation, both WHEY and SOY, would attenuate the increase in inflammation (IL-6) associated with the dialysis procedure. Furthermore, we expected to see a greater attenuation in the SOY group due to bioactive components of soy protein.

Subject Recruitment, Screening, and Selection. Individuals with CKD receiving hemodialysis treatment at the Champaign-Urbana Dialysis Center in Champaign, IL (n=32) and the Oak Park Dialysis Clinic in Oak Park, IL (n=12) were recruited at their respective clinics. Patients that met the following criteria were enrolled in the study: 1) Subjects must receive hemodialysis treatment at least 3 days per week. 2) Subjects must be ≥ 30 years of age. 3) Subjects must not have congestive heart failure or chronic obstructive pulmonary disease 4) Subjects must have been receiving dialysis treatment for ≥ 3 months as disruption in metabolic factors is related to duration of dialysis treatment. A Health and Medical History Questionnaire was administered during the screening, and informed consent was obtained from each participant. After consent was obtained, participants were randomly assigned to one of three groups: whey protein

(WHEY), soy protein (SOY), or placebo/control (CON). All research protocols were conducted with the approval of the University of Illinois Institutional Review Board. Information on participant recruitment, enrollment, and study completion is provided in Figure 6.1. Methods

Blood Collection. The study protocol consisted of two blood draws per day on two separate days, one week apart (Figure 6.2). The first draw was taken immediately after the initiation of dialysis and then again 3 hours into dialysis treatment. On Day 1, participants did not receive a study beverage (control day) but had the two blood draws as described above. The first draw on Day 1 is referred to as “baseline”. On Day 2, one week following Day 1, each participant received the study beverage to which they had been randomly assigned and consumed the beverage immediately prior to the initiation of dialysis treatment. After they consumed the beverage, blood was collected at the two time points described above. Therefore, each patient had one control day and one treatment day to allow for comparison of the treatment effect to their normal inflammatory response to dialysis treatment. Plasma was collected by centrifugation and divided into 1ml aliquots and stored at -80oC until analyzed.

Blood collections were not performed on Mondays or Tuesdays to control for the extra day between treatments that occurs during the weekend. For example, blood would not be drawn for a Monday/Wednesday/Friday scheduled patient on a Monday as they would not have

dialyzed since Friday (two day lapse). For a Tuesday/Thursday/Saturday scheduled patient, blood would not be drawn during their treatment on Tuesday for the same reason. All blood chemistries and data analysis were performed in Champaign, IL regardless of collection site.

Intervention: Protein Supplementation. Participants were given 30 grams of either a whey protein beverage, soy protein beverage, or placebo beverage immediately before dialysis

treatment as described above. Components of the protein beverages are listed in Table 6.1. The whey beverage (True Protein, Inc., Oceanside, CA) contained 27 grams of cold-filtered whey protein isolate per 30 gram serving, with 110 calories per serving. The soy protein supplement (Solae, Gibson City, IL) contained 27 grams of soy protein isolate (Supro 670) per 32 gram serving, with 120 calories and 1.5 grams of fat. This soy protein isolate provided 40 milligrams of isoflavones per serving (12 mg daidzein, 22 mg genistein, 6 mg glycitein). Supplements were matched for protein amount, resulting in a slightly higher total gram amount of the soy protein supplement. The powders were mixed with a flavor pack containing natural/artificial flavors and colors, sucralose, and acesulfame k, and were prepared with 4-6 ounces of water for

consumption. Both protein powders provided sodium, iron, phosphorus, calcium, and other nutrients that are of concern for dialysis patients; however, the levels provided in this beverage were lower than those in a commercially available renal-specific formulation at the same level of protein (Nepro, Abbott Nutrition, Columbus, OH). Patients in the CON group received 4-6 ounces of non-caloric Crystal Light prepared according to package directions (Kraft, Northfield, IL).

Blood Chemistry. Circulating levels of the inflammatory marker interleukin-6 were measured at all four study time points using commercially available ELISA kits (R&D Biosystems, Minneapolis, MN). In addition, blood collected at the dialysis clinic during normally scheduled blood draws was assessed for standard clinical lab parameters (plasma albumin, phosphorus, calcium, etc.) by a Spectra Laboratories, a renal specific laboratory service provider (Rockleigh,

NJ). Reported transferrin saturation is the ratio of serum iron and total iron-binding capacity, multiplied by 100.

Anthropometric Measures. Barefoot standing height was measured to the nearest 0.1 cm with a stadiometer and body weight was measured on a balance scale with shoes and superfluous outer garments removed. All measurements were taken in duplicate and averaged.

Statistical Analysis. Data presented are mean ± SEM unless otherwise noted, and significance was considered as p<0.05. Change in IL-6 levels was calculated using the following method: Δ Day 1 = Time Point 2 on Day 1 (3 hours into dialysis) minus Time Point 1 on Day 1 (start of dialysis); Δ Day 1 represents the change in IL-6 during dialysis with no study beverage. Δ Day 2 = Time Point 2 on Day 2 (3 hours into dialysis) minus Time Point 1 on Day 2 (start of

dialysis/consumed study beverage); Δ Day 2 represents the change in IL-6 during dialysis with the study beverage.

Using the Δ values calculated above, Δ Day 2 minus Δ Day 1 represents the difference of the change in IL-6 during dialysis after consuming the study beverage compared to the

difference in change of IL-6 during dialysis with no study beverage

Distribution statistics were calculated to determine whether assumptions of normality were met (including skewness and kurtosis between -2 and 2). For some analyses, data from WHEY and SOY were combined to look at the effects of protein supplementation, regardless of source; this combined group assessing general protein intake is labeled PRO. Correlation

analysis was used to identify relationships between selected variables of interest. Multiple linear regression analysis was used to determine predictors of intradialytic change in plasma IL-6 levels. For linear regression analysis gender and diabetes status were coded as follows: male=1, female=2; diabetes=1, no diabetes=2. All analyses were conducted using SPSS v.17 (Chicago, IL).

. The “Δ Day 2 minus Δ Day 1” variable was analyzed using ANOVA to compare means among the three treatment groups. Time point 1 and 3 on each day and Δ values for Day 1 and Day 2 were analyzed by paired samples t-test.

Patient Characteristics. A total of 44 hemodialysis patients (WHEY, n=20; SOY, n=15; CON, n=9) completed this study. Patient characteristics are listed in Table 6.2. There were no

significant group differences for age, weight, BMI, albumin levels, transferrin saturation,

hemoglobin, hematocrit, or iron. According to the National Kidney Foundation Kidney Disease Outcomes Quality Initiative (KDOQI™), patients were in normal range for transferrin saturation (20-50%), albumin (3.5-5.5 g/dL), and iron (30-160 mcg/dL) and were slightly low compared to the reference range for hematocrit at 37-47% and hemoglobin at 12-16 g/dL.

Results

Intradialytic Change in Plasma IL-6 Concentration (Three Group Analysis). Intradialytic changes in plasma IL-6 concentration are shown for WHEY (n=20), SOY (n=15), and CON (n=9) in Figure 6.3. All IL-6 values are expressed in ng/mL. On Day 1, plasma IL-6 increased by 2.94±1.2, 6.92±1.4, 3.00±1.2 in WHEY, SOY, and CON, respectively. The increase within each group was significant (p<0.01), but there were no between-group differences Δ Day 1. On Day 2, IL-6 increased 2.68±0.86, 3.45±1.4, 5.24±2.1 for WHEY, SOY and CON; the increase within each group was significant (p<0.05). The “Δ Day 2 minus Δ Day 1” mean values were - 1.28±1.2, -3.47±1.6, and 2.24±1.6 for WHEY, SOY, and CON with a trend toward a significant time by treatment interaction (p=0.065). Comparison of within group differences between Δ Day 1 and Δ Day 2 showed SOY was significantly lower from Day 2 to Day 1 (p<0.05) while the differences within WHEY and CON between Δ Day 2 and Δ Day 1 were not significant.

Intradialytic Changes in Plasma IL-6 Concentration (Two Group Analysis). Intradialytic changes in plasma IL-6 concentrations are shown for PRO (n=35) and CON (n=9) in Figure 6.3. On Day 1, IL-6 increased by 5.23±0.94 in PRO and increased 3.01±1.22 in CON; the increase within each group was significant, but there were no between-group differences for change in IL-6 under the condition of no beverage on Day 1. On Day 2, PRO increased by 3.01±0.79 while CON increased by 5.24±2.1; again, the increase within each group was significant and there were no between group differences.

There was a significant time by treatment interaction for “Δ Day 2 minus Δ Day 1”; mean values were significantly lower in PRO (-2.22±0.98) compared to CON (2.23±1.6)

by around 2 pg/mL. Although the “Δ Day 2 minus Δ Day 1” variable for the CON was positive (2.23±1.6), paired sample t-test analysis confirmed the within-group difference between Δ Day 1 and Δ Day 2 was not significantly different for CON between no beverage day and beverage day (p=0.196); paired sample t-test analysis of PRO for Δ Day 2-Δ Day 1 was significant (p<0.05), suggesting an effect of treatment on Day 2. Furthermore, ANOVA of the mean “Δ Day 2 minus Δ Day 1” as described above confirmed the significant time by treatment effect when mean differences in change were compared between CON and PRO.

Relationship of IL-6 to Albumin, Serum Iron, and Transferrin Saturation. We found a significant inverse relationship between plasma albumin levels and both baseline IL-6

concentration (r=-0.451, p<0.01) (Figur e 6.4) and change in IL-6 on Day 1 (=-0.484, p<0.01) (Figur e 6.5). There was a significant negative association between plasma transferrin saturation and both baseline IL-6 concentration (r=-0.348, p<0.05) (Figur e 6.6) and change in IL-6 on Day 1 (r=-0.378, p<0.05) (Figur e 6.7); serum iron was also inversely related to baseline IL-6 (r=- .356, p<0.05) and change in IL-6 on Day 1 (r=-0.407, p<0.01). However, only baseline IL-6 was predictive of change in IL-6 in a model that included gender, time on dialysis, diabetes status, albumin, and transferrin saturation (Table 6.3).

We found that oral administration of 27 grams of protein immediately before dialysis treatment attenuates the increase in IL-6 associated with treatment, the first study to suggest the potential anti-inflammatory effects of intradialytic protein supplementation. In addition, baseline IL-6 and the intradialytic change in IL-6 were inversely related to circulating albumin, transferrin saturation, and iron.

Discussion

CKD is a chronic inflammatory condition, as reflected in dialysis patients by elevated circulating levels of acute phase proteins such as CRP, and pro-inflammatory cytokines such as IL-6, and TNF-a56-59. The dialysis procedure also induces an acute inflammatory response230, 233; we were able to confirm the increase in IL-6 that occurs during dialysis treatment by showing an increase in IL-6 on Day 1 when no treatment was administered, although the increase may be due in part to hemoconcentration occurring during treatment. However, by measuring each person under the same dialysis conditions on both days, this effect was minimized and allowed

for a relative comparison of the effect of protein supplementation on inflammation during dialysis.

In a study by Caglar and colleagues, IL-6 levels in hemodialysis patients peaked during a 2-hour post dialysis phase233, but we chose to collect blood at 3 hours into dialysis treatment in order to minimize the burden of the patient to remain in the clinic for several hours after completing dialysis treatment. While it is unclear what the effects of protein supplementation would be in this post-dialysis phase, we were able to demonstrate an effect of protein at the three hour time point. Furthermore, our study had a larger number of subjects (n=44) compared to the Caglar study (n=9) that may have allowed for earlier detection of the increase in IL-6 that occurs in conjunction with dialysis treatment.

We were able to show that intradialytic protein supplementation attenuates the increase in inflammation associated with the dialysis procedure. We compared the change in IL-6 over a dialysis session in patients on both a control day and a treatment day, allowing for better control over the variability of the inflammatory response in these patients. Although we were able to find an acute benefit, the beneficial long term effects of this acute reduction are unknown. Bossola et al reported time on dialysis was associated with improvements in nutritional status and a trend towards a reduction in inflammatory variables over a three year period107, which seems contrary to current evidence considering the acute pro-inflammatory and catabolic effects associated with dialysis treatment compounded over time. The authors concluded time on dialysis was not necessarily associated with a decline in nutritional status or increase in

inflammation; however, patients routinely received an intradialytic meal containing 25-30 grams of protein. Because this meal was given as part of routine care, they did not consider it to be part of the study protocol. This suggests routine intradialytic protein and/or calorie intake may be responsible for attenuating the decline in nutritional status and increased inflammatory state associated with long-term dialysis treatment. However, this study lacked a control group and failed to provide information on subject compliance for this intradialytic meal, and more well- controlled studies are clearly needed to see if the acute reduction shown in our current study would translate into long term beneficial outcomes. Furthermore, the effects of protein supplementation on functional outcomes related to inflammation in CKD patients, including cardiovascular risk and bone disease, warrant further investigation.

Although we demonstrated the attenuation of the increase in IL-6 in the protein combined protein group, we saw a trend for a time by treatment interaction when all three groups were considered (p=0.065). Although the overall time by treatment interaction was not significant, the within-group difference for SOY was significantly lower for Δ Day 2 compared to Δ Day 1 (p<0.05). Therefore, it may be possible that the source of the protein may be a factor in determining the intradialytic inflammatory response to protein supplementation. Fanti and colleagues demonstrated a marked inverse relationship between plasma CRP levels and the change in plasma isoflavone concentration measured before and after an 8-week supplementation with 25 grams of intradialytic soy protein118; they also noted a significant positive correlation between change in plasma isoflavone concentration and plasma albumin levels measured at the end of the study. A milk protein supplement was used as the control in this study, and the authors did not report on the relationship between milk protein intake and plasma albumin levels albumin, making it difficult to determine if the association of isoflavones and albumin was due to the protein component of the supplement or a function of the bioactive isoflavones found in soy. However, this study and others suggest a possible relationship between soy protein consumption and inflammation during dialysis treatment.

Protein supplementation during dialysis treatment, regardless of the protein source, will increase substrate availability, and this may help explain the modest effect of whey protein when the three groups were analyzed separately. Compared to soy protein, whey protein contains higher amounts of the branch-chain amino acids leucine, isoleucine, and valine. Leucine

concentrations are especially important during a highly catabolic condition, and high amounts of leucine have been shown to potently stimulate muscle protein synthesis and inhibit protein breakdown in skeletal muscle and liver (reviewed in128). Therefore, intradialytic whey protein supplementation may potentially mediate the inflammatory state by reducing catabolism through increased availability of amino acids. Intradialytic protein supplementation may be particularly